One evening last October, a University of Texas at Austin astronomer named Gary Hill stood behind a lectern at Miss Hattie’s Café and Saloon, which occupies a restored nineteenth- century bank building in downtown San Angelo, and cheerfully proclaimed his ignorance. “We really don’t understand the universe,” he said. “We thought we did, but it turns out we only understand about four percent of it.”

A dozen or so people, among them businessmen, the editor of the local newspaper, a school librarian, and a couple of college professors, had assembled at Miss Hattie’s, where the lace curtains and rose-print wallpaper harked back to a time when the universe was no larger than our own galaxy and Newton’s laws seemed to explain it and a tunnel linked the building where we sat to a nearby bordello. There was something old-fashioned too about the fact that a bespectacled, British-born scientist had traveled from the state capital to give a talk to the curious and that the curious had turned out to hear what he had to say. His subject, on the other hand, was the very future of cosmology and physics and how they might be affected by one telescope in particular, located 212 miles farther west.

Galileo stuck lenses onto either end of an organ pipe; today’s research telescopes, while considerably more elaborate, still perform the same fundamental task of collecting and focusing light. It’s all astronomers have to go on: electromagnetic radiation from distant objects, whether it arrives in the form of X-rays or visible light or radio waves. “We’re detectives, but we can only use what light will give us,” Hill had said to me earlier that day. “So we get fairly ingenious in the ways we analyze light to look for clues.” They rely, for instance, on spectroscopy, the process of separating light emitted by an object in space into its component wavelengths, as a prism does, then analyzing those components. And they invent new tools to analyze the light. To probe deeper and deeper into space, scientists must design better and better detectors, sensitive to the faintest of emissions.

Such instruments don’t come cheaply, which is why Hill and his colleague Karl Gebhardt have periodically taken to the road over the past three and a half years: They’ve been promoting an ambitious $34 million overhaul of a telescope at UT’s McDonald Observatory, in the Davis Mountains of West Texas. Speaking to potential donors in Houston or a luncheon group in Abilene, they’ve been publicizing an endeavor called HETDEX, or the Hobby-Eberly Telescope Dark Energy Experiment, the aim of which is to help attack what some have labeled one of the most important problems in science.

Hill grew up in England but left to go to where the telescopes were, first as a graduate student in Hawaii, then as a postdoctoral researcher in West Texas, in 1988. Finding that certain instruments at the observatory weren’t sensitive enough, he designed and built a new spectrograph on the cheap, still in use today. He is now the observatory’s chief astronomer. At Miss Hattie’s he applied his knack for innovative thinking to the problem of business attire—he wore a striped green shirt and a pink tie—and as he spoke, he grinned and nodded infectiously. “We have a huge opportunity to lay the groundwork in Texas for understanding how the universe has changed through time,” he said. He outlined the goals of the experiment: to conduct the largest survey of other galaxies ever completed and to use that information to measure how the scale of the universe has evolved—and to reinvent the telescope in the process. By doing so, Hill, Gebhardt, and their collaborators hope to better understand what astronomers call dark energy, though no one really knows what the term means: “Dark energy” is a label for a mystery. “The thing is,” Hill told his audience, “it may not be dark and it may not be energy.”

After the presentation had ended and most of the audience had departed, Hill was subjected to a more rigorous interrogation at dinner from Ken Gunter, a tall, poker-faced San Angelo businessman and a member of the McDonald’s Board of Visitors, a statewide group of observatory supporters. This turned into a debate between the gruff West Texan and the polite but impassioned British scientist, while half a dozen others at the table looked on. Though he supports the experiment, Gunter was skeptical as to whether he could raise funds for it. “What is the pragmatic end product, except exciting a bunch of astronomy Ph.D.’s in Austin? Give me something I can identify with. If you want to raise some honest-to-God money, you better start raising some honest-to-God tie back to medical or energy.” That, said Hill, wasn’t the point. The most practical argument he could make for the telescope was that it might excite students in a country where science education was failing and draw better faculty to the university. Gunter seemed unconvinced. “My sense is that most people don’t give two hoots in hell whether the universe is expanding or contracting or moving sideways!” he said, and drew out a cigar.

Earlier that day I’d ridden in a rented suburban with Hill; David Lambert, the director of the observatory; and Joel Barna, its development director, from Midland to Abilene to San Angelo. The flat landscape was staked by the technology of energy production: Near Midland the pump jacks kowtowed to the brush, while farther south a line of soaring silver windmills receded toward the horizon. (One question often posed by lay audiences in Texas, Hill told me, is “How can we harness dark energy for human use?”) It was a warm, hazy morning, and as we’d all risen early to catch a seven o’clock flight from Austin to Midland, a soporific air had fallen over the car, leaving me in a state of sleepy wonderment at one of astronomy’s fundamental principles, that light can ripple for billions of years through the vast universe and collide with nothing else during that unfathomably long journey, reaching our planet with information about its place of origin. How can that be, I asked. “There’s a lot of empty space out there,” Hill said. “The fraction of the area of the sky with stuff is not very great, so the chances of a photon of light hitting anything along its route other than your telescope are actually very small.”

I fell silent. All that void. A little while later I asked something else, then mulled over the answer, and that’s pretty much how the ride went, as I surfaced every so often with another question for the patient professors, then let my head spin for a while. It’s humbling to stumble up against the edifice of astronomy, the massive and intricate body of knowledge humans have built up over the millennia, and then all the more humbling—if somehow also reassuring—to contemplate how much more we don’t know. And how much we can’t possibly see, since only 4 percent of the stuff in the universe is thought to be visible matter. That includes you and me and microorganisms and other galaxies, at least 100 billion of them. As one scientist put it recently, “we’re just a bit of pollution,” while most of the universe is made of something else.

“Nature and nature’s laws lay hid in night” goes the famous epitaph Alexander Pope composed for Isaac Newton. “God said, ‘Let Newton be!’ and all was light.” But as it turns out, almost all is, in fact, dark, and a crucial portion of those laws remain hidden. Theorists first ventured into the “dark” to resolve a problem concerning the motions of galaxies and the stars within them. The trouble was, there just weren’t enough visible stars out there. Matter far away from the center of a galaxy tended to move much faster than could be explained by the net gravitational attraction of all the visible matter. So in order to save Newton’s laws of gravity, astronomers invented a new type of matter that was dark but plentiful: dark matter. The universe is littered with it, they concluded. Which might seem like cheating—the invisible check is in the mail—yet the theory has helped explain the motions of other cosmic entities, and there are high hopes that dark matter will be detected in experiments this year at CERN, the particle physics laboratory in Geneva, Switzerland.

Then scientists discovered another anomaly, this one not limited to objects like galaxies but pertaining to the entire cosmos. And they came up with something even stranger to explain it. The problem was the way in which the universe was growing. Until the twentieth century, the universe was generally thought to be static, neither expanding nor contracting. But along came Edwin Hubble (who, late bloomers take note, taught high school and coached basketball in Indiana before returning to school for his Ph.D.). In the twenties he made measurements of how far away galaxies were and how fast they were moving, and he discovered that the farther away they were, the faster they were receding. The universe, this implied, was expanding, everything moving away from everything else. (Albeit with negligible impact locally. In Annie Hall, Alvy Singer’s mother is basically correct when she tells her existentially morose son that “Brooklyn is not expanding.”) The concept of an expanding universe in turn suggested that at one time, everything was closer together. Thus Hubble’s discoveries were a prod to the subsequent development of the big bang theory.

Meanwhile, a decade before Hubble, Einstein had revised Newton’s idea of gravity—an attractive force between massive objects—with his theory of general relativity, in which gravity depends on the curvature of four-dimensional space-time. The twentieth century saw the Newtonian model of a fixed, orderly universe supplanted by that of a warping, expanding space with unimaginably chaotic origins. This universe had a beginning and would perhaps come to an end as well: If the expansion of space was indeed due to the propulsion from the big bang, then it stood to reason that because the gravitational pull of all the matter in the universe would be retarding that outward growth, one of three things would ultimately happen: Gravity would win out and the universe would collapse, the initial propulsion from the big bang would win and the universe would expand forever, or—and this is the most accepted case—the two would be perfectly balanced.

But in 1998, two groups of researchers independently arrived at a surprising result, so surprising that members of each originally believed that they had made a mistake. Both teams had been studying a class of exploding stars called type 1a supernovae, measuring their distances and speeds. What each team found was that the most distant supernovae were fainter than expected, fainter than they would have appeared in a decelerating or even a coasting universe. The expansion of the universe, this meant, seemed not to be slowing down but rather speeding up. And no one had a good explanation for it. What phenomenon could possibly be pushing space outward?

“Theorists have been guiding observers for ten thousand years,” said Edward “Rocky” Kolb, a professor of astronomy and astrophysics at the University of Chicago. “Now they’ve observed something that we don’t understand. We desperately need a better idea of what’s going on.” The notion that some form of mass energy might pervade all of space dates back to Einstein, but it gained widespread acceptance only in response to the observed acceleration of the universe. But what sort of energy? Has it changed over time? Or could it be that there really is no such thing as dark energy and that our understanding of gravity is actually wrong? These questions demanded new experiments, which in turn would require new telescopes or modifications to existing ones.

Karl Gebhardt arrived at the University of Texas in 2000, an expert on nearby galaxies and black holes. But in 2004 he attended a meeting on the future of U.S. telescopes, where much of the talk was about dark energy.

“It really hit home: Everyone is pushing on this,” he told me when I visited him in his office. “All the dark energy missions were beginning to take root, and I said, ‘Hey, look, I think we can do this at Texas.’ I came back from the meeting all excited. So one day in the hall I was talking to Gary Hill about it, and I said, ‘Can we do this? Can we study dark energy?’ It turned out he was working on an instrument design with Phillip [MacQueen, the observatory’s chief scientist and a senior researcher at UT-Austin], and we decided we can make a really nice instrument to look at the problem.”

For the next few weeks, Gebhardt and Hill traded ideas about how to proceed, and MacQueen and Hill consulted about whether the instrument they’d conceived would be up to the task. “We went back and forth for a few months until we came up with a nice design,” Gebhardt said. “But it’s easy to come up with designs. The hard part is finding the money and the telescope to do it on.” Here, though, the team had two advantages: the Hobby-Eberly Telescope and a recent grant from the Air Force.

Not long afterward, they began trying to raise the $34 million, more than the cost of the original telescope. The financial end was both a burden and a potential edge: The Texas scientists would not be bound by the same sort of bureaucratic limitations that they might have encountered using a federally administered facility or depending heavily on government grants. “The National Science Foundation won’t go off the beaten path until they’ve beaten the path,” said Gary Bernstein, a professor of physics and astronomy at the University of Pennsylvania. “If you have your own facility and can do what you want to do, you have more freedom.” Or as Hill said to me, “In Texas, if you have the idea and you have the money, you can do it.” And the $34 million, while hardly a small sum, was much lower than the costs of other proposed dark-energy experiments.

A plan to put together a world-class instrument on the cheap might sound an alarm in the mind of anyone familiar with the observatory’s history, for a cost-saving design had been trumpeted once before, in building the Hobby-Eberly Telescope itself. Though less expensive than other telescopes of comparable size, the HET had suffered technical problems for several years after it was dedicated, in 1997. While it was one of the largest telescopes in the world, it had gained a reputation for poor image quality. Now, at last, it was functioning properly, and here came some upstart astronomers with a proposal to chop off its top half and replace the detector with a novel apparatus. Could grand ambitions get the better of the observatory a second time?

If the scientists themselves had such doubts, they kept them quiet. Their concerns were more concrete: Raise the money and get results before anyone else. Among the other proposed dark-energy experiments, only one will employ the same technique as HETDEX, and that’s a collaboration between scientists in Japan and the United States to build something called the Wide-Field Multi-Object Spectrograph, which would be installed on the Japanese-operated Subaru Telescope, in Hawaii. “This is a strange game, in that you don’t know exactly how much data you have to take before you can make a great discovery,” Kolb said. “Timing is of the essence. It’s sort of a race between the Japanese and Texans. This is such a hot topic that many people are interested in it. There is the potential of great discoveries to be made, and so you just can’t sit on it and wait.”

The HETDEX researchers, led by the Texas group and further supported by two universities in Germany and the consortium that operates the HET, are the mavericks of the dark-energy industry, relying on a small number of people and a relatively low-cost instrument. So far they’ve raised $20.1 million—enough to build a prototype for the elaborate new spectrograph and to rebuild the HET to give it a much wider field of view—and they’re gunning to finish their upgrade and their observations by 2013. “We are dwarfed by these other teams,” Gebhardt said. “They have the ability to go out there and barrage everybody at conferences. We can’t do that. We are working all-out just to get our instrument built. But it never hurts to be the underdog. I think people are a little scared of HETDEX because we’re going to finish soon now. We are really moving.”

If you were to look at a nighttime satellite image of Texas, you would see large splotches of light produced by Dallas—Fort Worth and Houston, smaller ones for Austin and San Antonio, and a grid of glowing points as you move west from Austin, the most prominent strand of them following Interstate 10. Gradually the lights trail away; in West Texas south of the interstate, the map goes almost completely dark. It’s one of the darkest spots in the continental United States, making it a natural place for an observatory.

The location was identified in the summer of 1932 by two astronomers who drove a Chevrolet nearly eight thousand miles—roughly one third of the earth’s circumference—all within Texas. They ranged from Galveston to El Paso and as far north as Amarillo, stopping to peer through a small telescope, make notes, and take photographs at prospective sites. An East Texas banker named William McDonald had bequeathed his fortune to the University of Texas so that it could build an astronomical observatory (for the stated purpose of “seeing closer the gates of heaven”), and the university, which had no astronomy department, contracted with scientists from the University of Chicago to supply the expertise—the first order of business being to choose a location. They selected the Davis Mountains, a choice approved by Otto Struve, the Russian émigré who became the observatory’s first director, after he spent several nights camping on Mount Locke with his wife.

Today, as you drive from the interstate to the town of Fort Davis and on to the observatory, the elevation rises, the land begins to buckle and swell, and having passed over miles of desert plains, you find yourself driving through cedar-sprinkled hills and finally up Mount Locke, where most of the observatory is located. At first you see two white domes resembling centurion helmets at the summit. They are the observatory’s older telescopes, the Struve telescope, or “the 82-inch,” after the diameter of its primary mirror, which was built in 1939, and the Harlan J. Smith 107-inch, completed in 1968. Each in its time was one of the most powerful telescopes in the world, only to be eclipsed, inevitably, by technology’s advance. (They are still in use, but they are no longer considered cutting-edge.) Higher up the hill a silver-mirrored dome resembling a giant buckyball comes into view: This, situated on an adjacent peak, is the Hobby-Eberly Telescope.

With its own water supply, fire marshal, staff homes, and lodge for visitors, formerly called the Transients Quarters (now the Astronomers’ Lodge), the observatory is its own little campus on a hill, one that abides by strange customs. People drive around at night with only their parking lights on, for instance, to keep things dark. It’s a quiet place, where an influential portion of the population sleeps during the day and works at night, and the sky commands attention—not just from astronomers but from anyone who finds herself on the mountain. The night I arrived at the observatory, a woman passing me on the road told me to be sure to take in the stunning sunset; the next night another person bid me look out at the storms to the west.

But there’s the sky we see, and then there’s the sky astronomers investigate. Enter through a side door of the HET building, and it’s as if you’ve stepped into a small manufacturing plant, with a concrete loading area and a flight of metal stairs leading to the site manager’s spare office. The manager in this case is Bob Calder, who worked for the Subaru telescope, in Hawaii, and the Smithsonian Astrophysical Observatory, in Cambridge, Massachusetts, before moving to Texas to oversee the operation and maintenance of the HET—though “maintenance” is perhaps too plain a word, too reminiscent of boilers or car engines, to convey the kind of care one has to take with a major telescope. Even cleaning is a delicate task. To focus on the very thin slices of sky required for astronomical work, the telescope’s large primary mirror must be absolutely smooth, free of any imperfections larger than one tenth the wavelength of the light hitting it. The aluminum coatings on the HET’s mirrors are the thickness of oil on water; they are cleaned on Mondays, Wednesdays, and Fridays by a four-person team that applies a carbon dioxide “snow” with a wand.

As site manager of the HET, Calder also greets the dignitaries and potential donors who come to tour the telescope, a duty that has, on occasion, served to remind him that he is not in Cambridge any longer. “We had a guy one time who tried to give me a wad of bills after the tour,” Calder recalled. “He goes, ‘Let me do something for your family.’ I suggested he donate to the observatory instead, and he said he was going to, but he still wanted to give me some money too.” (Calder declined the gift.)

He escorted me down through the control room, where the telescope is manipulated via computers and electronics, and inside the dome. It is a sort of cross between a science cathedral and a science factory, given over to one giant piece of industrial equipment aimed at the heavens: the HET. While it calls to mind some sort of futuristic vehicle poised to be launched into space, the telescope actually rotates on a fixed base, letting space come to it. Ninety-one hexagonal pieces form the primary mirror, a silver dish inside a steel cage; the light from the night sky hits this and then is relayed to an instrument above the mirror called the tracker. Itself a complex piece of machinery that took two years to build, the tracker can follow the image of a particular object as that image crosses the mirror. (Due to the rotation of the earth, the sky and its contents shift, relative to the position of the telescope, over the course of the night.)

Calder flipped a switch, engaging the air bearings below the telescope, and like a big hovercraft, it rose a few inches and then began to rotate. Moments later, a recording of squawking birds began to play, which is supposed to ward off real birds that might fly in through the ventilation shafts or through the top of the dome. Speaking over the sounds of the machinery and the squawking, Calder summarized the changes planned for the HET. “There’s one camera up there right now,” he said. “That whole top end is going to be replaced with a new top end that supports on the order of one hundred and fifty cameras, one hundred and fifty spectrographs.” Those spectrographs, he told me, would be deployed, over the course of 140 nights, to record one million galaxies.

Later that evening I crouched on a platform below the Smith telescope, where Hill, MacQueen, and graduate student Josh Adams were bustling about a prototype of the instrument that is slated to be put on the HET. The instrument is called VIRUS, for Visible Integral-field Replicable Unit Spectrograph, and it will incorporate more than 145 copies of a simple spectrograph, each of which registers data from at least 246 fibers arrayed within something called an integral-field unit. Scattered on the floor of the platform were scissors and cables and screws and Allen wrenches, as if the scientists were fixing a motorcycle rather than testing a highly engineered piece of equipment, which was contained within an irregularly shaped black box about four feet wide and attached to the underside of the telescope’s shaft by a metal harness. I watched as they checked the connections between the telescope and the box, cooled a crucial part with liquid nitrogen, and installed a glass plate, all in preparation for an observing run.

“There are four big links in the chain of observational astronomy,” MacQueen explained as he connected the liquid nitrogen tank to the device. “Collecting the light, processing the light, detecting the light, and all the software side of analyzing the light.” HETDEX will, if all goes smoothly, increase by about thirty-fold the area of sky that can be observed at any one time, speeding up the first three steps and thereby transforming the HET into a far more efficient and comprehensive cartographer of the skies. Rather than pointing at and shooting known objects, the upgrade will allow the research team to methodically obtain spectra from a giant celestial swath, taking information from more than 40,000 pixel-like sections of sky per exposure. Those will be captured by an 800-megapixel camera whose technology is similar to that of any vacationer’s digital camera. “This is just a super-duper digital camera of sorts,” MacQueen said.

Yet this is a camera that takes pictures of the remote past. Imagine a man in California trying to construct a map of the United States, using only the information brought to him by slow-moving (but long-lived) ants, creeping along at a hundred yards per year. The ants from New Jersey would have departed more than 50,000 years ago, before humans arrived in North America; the Midwestern ants some tens of thousands of years ago; the central California ants just a few years ago—so the map would record the East Coast as wild and uninhabited while the West Coast would contain highways and burger joints. This is the sort of map astronomers construct of the universe: The light that arrives from greater distances relays information about earlier eras.

If the big-bang model of the universe is correct, though, we cannot receive any messages from the earliest time periods, any more than our Californian mapmaker could learn about France. For some 380,000 years after the big bang, the universe was so hot that particles couldn’t form atoms; electrons whizzed around every which way, and light couldn’t travel any distance without colliding with those electrons. While this early fireball of a universe remains opaque to us, it is theorized that there must have been random variations in how densely its particles were distributed. Some areas were more dense than average, creating hot spots that drove particles out to cooler regions, which in turn caused waves to travel at the speed of sound. The result was a series of ripples through the sea of particles.

When the universe cooled down enough for atoms to form and light to escape, the oscillations ceased. At that particular moment in the history of the cosmos, the ripples were frozen into the density distribution of matter—so that now when astronomers map the heavens, they find the imprints of the early ripples in the way that galaxies are clustered. (It’s as if you drew a pattern on a balloon and then inflated it. As the balloon expanded, the pattern would persist though its scale would change.) These imprints are what the HETDEX team plans to measure, to learn how they might have grown over time. By comparing the traces of the oscillations at different epochs, both with one another and with a pattern already discovered in something called the cosmic microwave background—which gives us a picture of the universe just after the light first escaped—they will be able to evaluate the changes in the scale of the universe and from that infer whether dark energy has made a constant contribution to the universe’s expansion or has varied over the eons.

Some cosmologists have questioned whether a concerted push to do dark-energy experiments would be detrimental to the wider field of astronomy, hindering discoveries in other areas and relegating too many scientists to supporting roles on large-scale endeavors. After all, maybe there isn’t any such thing as dark energy. Maybe no one really understands how gravity works. Or maybe the universe is not as homogenous as theorists have assumed. Maybe “dark energy” is the wrong way to frame the problem.

But even the possibility that dark energy might be a mistaken notion is no cause for pessimism, according to Gebhardt. “That would be the most exciting answer—if we’re just so completely out to lunch,” he said. “It means there is something fundamental about the nature of our universe that we don’t understand. So whatever the answer is, it’s going to be a fundamental shift in our understanding. In other words, it’s not going to be a boring answer, and who knows where it will lead?”